A common theme in the pathogenesis of the neurodegenerative diseases is that a widely expressed mutant protein can cause selective degeneration of a subset of neurons in the brain. In most of the cases, the molecular and cellular mechanisms underlying such selective neurodegeneration remain unclear. Using Huntington's disease (HD) as a model, we will investigate whether selective degeneration of the striatal and cortical neurons in HD requires mutant Hungtingtin (mHt) toxicities originated from outside these neurons (non-cell-autonomous toxicities). In our preliminary studies, we have developed two conditional mouse models in which mHt expression can either be activated or inactivated in specific neuronal populations. We propose to use these mouse models to achieve the following three Specific Aims: Specific Aim 1. Phenotypic analyses of the cortical activation mouse model and the whole brain activation mouse model to determine whether non-cell autonomous toxicities are involved in dysfunction and degeneration of the vulnerable cortical and striatal neurons. Specific Aim 2. Phenotypic analyses of the striatal activation mouse model to determine if the non-cellautonomous toxicities are involved in dysfunction and degeneration of the striatal and cortical neurons. Specific Aim 3. Generation of a full length mHt cortical-inactivation mouse model to study whether non-cell autonomous toxicities are required in disease pathogenesis in a full length mHt mouse model. In summary, we propose to use a novel series of conditional mouse models of HD to study the role of neuronal circuitry in HD pathogenesis. Our studies may reveal that mHt toxicities originated from outside the most vulnerable neurons may be required for their dysfunction and degeneration. Identification of such toxicities will have important implications for developing therapeutics for HD as well as for other neurodegenerative diseases.